[R-390] A 6kc. mechanical filter for a R-390a?

[Charles Steinmetz]

The original Collins filters (the type used in the 390A) are 
self-terminated and are designed to be fed from a Hi-Z source and 
terminated by a Hi-Z load (50k ohms or greater in parallel with 130pF). 
  Here is what a contemporary Collins applications note said:

> THE DESIGN OF CIRCUITS employing Mechanical Filters is relatively simple,
> since no special matching networks are normal]y required. Being internally
> terminated, the filters need only a high-resistance termination (50,000 ohms
> or greater) at either end together with the capacity (approximately 130 pf)
> required to resonate filter input and output at the center frequency.
>
> THIS HIGH RESISTANCE is readily obtained by driving the Filter with a
> pentode tube (effectively a constant current generator) and terminating it
> into a vacuum tube grid. It was this usage that led to the use of the term
> "transfer impedance" in specifying the effect of a Mechanical Filter on the
> gain of a given circuit. The transfer impedance is the ratio of the input
> current to the output voltage, so the over-all gain of an amplifier stage
> with a Mechanical Filter following the amplifier tube is simply equal to the
> transconductance of the tube times the transfer impedance.

Many of the newer rectangular filters are designed to be terminated at 
both ends with 2k ohms in parallel with 30pF.  I am attaching a 
schematic of the Collins test circuit below.  Your filter is most likely 
of this type.

Best regards,

Charles



Al S. wrote:

> The TM 11-856a Army technical manual for the R-390a says the input and
> output impedances of the filters are in excess of 20k ohms and resonating
> capacitances of between 120 to 140 pf. How to measure the impedances
> accurately.  Maybe its a known value.
>
> Al
>
>
> Alan Victor wrote:
>
>> First you need to establish the current terminating Z for the mechanical
>> filters in the 390A. Then you need to know the required termination Z for
>> the ones you wish to install. Designing a matching system to handle this
>> problem is pretty straight forward. I do not believe the current 390A
>> mechanical filters are that high a Z. At least the ones I have worked with
>> at 455 kHz and built some receivers around were only about 3k//?? Cpf. I
>> would have to look them up in my notes.
>>
>> Alan W4AMV


-------------- next part --------------
A non-text attachment was scrubbed...
Name: Mechanical_filter_test_circuit.png
Type: image/png
Size: 14171 bytes
Desc: not available
URL: <http://mailman.qth.net/pipermail/r-390/attachments/20170616/40c3c61b/attachment.png>